EP0002232B1 - System and method for monitoring the operation of an apparatus - Google Patents

System and method for monitoring the operation of an apparatus Download PDF

Info

Publication number
EP0002232B1
EP0002232B1 EP78101434A EP78101434A EP0002232B1 EP 0002232 B1 EP0002232 B1 EP 0002232B1 EP 78101434 A EP78101434 A EP 78101434A EP 78101434 A EP78101434 A EP 78101434A EP 0002232 B1 EP0002232 B1 EP 0002232B1
Authority
EP
European Patent Office
Prior art keywords
signal
monitoring device
alarm
electrical signals
signals
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired
Application number
EP78101434A
Other languages
German (de)
French (fr)
Other versions
EP0002232A3 (en
EP0002232A2 (en
Inventor
Robert Samuel Morrow
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
IRD Mechanalysis Inc
Original Assignee
IRD Mechanalysis Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by IRD Mechanalysis Inc filed Critical IRD Mechanalysis Inc
Publication of EP0002232A2 publication Critical patent/EP0002232A2/en
Publication of EP0002232A3 publication Critical patent/EP0002232A3/en
Application granted granted Critical
Publication of EP0002232B1 publication Critical patent/EP0002232B1/en
Expired legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07CTIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C3/00Registering or indicating the condition or the working of machines or other apparatus, other than vehicles

Definitions

  • the present invention is particularly adapted for use in monitoring vibrations produced by rotating or other types of machinery in a complete industrial installation, such as refinery.
  • a complete industrial installation such as refinery.
  • Computer systems generally include disc memory for data storage, CRT terminals for display of data and line printers for hard copy of data. As a result, they require a relatively large capital investment. While simple data loggers are relatively inexpensive, they offer simple functions only such as logging data and comparing the data to setpoints.
  • a system for monitoring the operation of an apparatus using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to apply respective electrical signals via multiplexing means to a computer apparatus, having memory means and read-out means for periodically storing at least selected ones of the electrical signals from each monitoring device in the memory means, the computer apparatus further including means for performing a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times in said memory means an alarm being initiated indicative of a significant trend.
  • the known system is exclusively based on a monitoring by way of a power spectral density analysis (PSD) which is a squared form of analysis which is generally non-linear.
  • PSD power spectral density analysis
  • a trend analysis is performed by taking samples of each of certain frequency intervals of the PSD plot and averaging the data over that particular interval for any one scan cycle to obtain an average value over that interval which is stored as historical data.
  • the historical data are then used for the trend analysis by taking the average value of the sample in any one frequency interval and doing a regression analysis to fit the best trend line of those data.
  • a trend line showing a significant slope in the way of increasing or decreasing value over a period of time indicates a significant trend and would alert the operator.
  • This system is one of the complex computer type systems. Nevertheless, only certain frequency intervals from the whole frequency range are monitored. With the known system it is necessary to store the average values for all frequency intervals separately. Though the operator is alerted and the appropriate corrective action is indicated, no information is given in respect of the urgency of such an action. Thus, if there are several alarm conditions at a time, he does not know the priority of the actions necessary.
  • the system for monitoring the operation of an apparatus using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to apply respective electrical signals via multiplexing means to a computer apparatus having read-out means and memory means for periodically storing at least selected ones of the electrical signals from each monitoring device, the computer apparatus further including means for performing a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times in said memory means, an alarm being initiated indicative of a significant trend, is characterised in that for a said monitoring device there is provided a means for generating said selected electrical signal corresponding to the overall amplitude of a plurality of signal amplitudes substantially corresponding to the frequency response range of a said monitoring device, that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that the said computer apparatus includes means for computing from the trend of said values the probable time to failure of the monitored apparatus and means for causing said read-out to provide information indicative of the probable time to failure.
  • a method for monitoring the operation of an apparatus using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to produce respective electrical signals including the steps of feeding each of said electrical signals from the respective monitoring devices to a computer apparatus, periodically storing at least selected ones of the electrical signals from each monitoring device performing in the computer apparatus a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times and initiating an alarm for a significant trend, is characterised in that said selected electrical signals correspond to the overall amplitude of a plurality of signal amplitudes substantially over the frequency response range of a said monitoring device, that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that from the trend of the said values the probable time to failure of the monitored apparatus is computed and that information indicative of the probable time to failure is provided.
  • a data monitoring system which does not require a large capital investment but which, nevertheless, is capable of reading out complete system information including a malfunction of any one of a number of different devices being monitored, the time to failure of any pieces of equipment being monitored, and an analysis of the input information.
  • the invention performs the functions of automatic channel data logging, frequency spectrum analysis, and vibration level trend prediction. Each of these functions additionally may be manually selected for each individual monitor or channel via front panel control.
  • a built-in system fault detection circuit is used which will respond to either an internal or circuit fault or to an external system alarm relay closure. Data readout is obtained via a self- contained dot-matrix printer assembly.
  • All functions of the monitoring system of the invention are under the control of an internal microcomputer which continuously samples data from a plurality of monitors.
  • vibration input signals are obtained directly from velocity pickups, self-amplified accelerometers, non-contact signal sensors or from accelerometer preamps.
  • direct current signals proportional to vibration level or amplitude and trip alarm signals are obtained from the monitors, these latter signals being derived by comparison of the actual vibration signal with reference signals proportional to preselected alarm and trip levels.
  • the system automatically indicates, via the computer read-out those channels which go into a trip condition within a preselected time span. That is, the time to failure is calculated and displayed via the print-out.
  • Each channel's "look ahead" time may be selected with a user-programmable jumber board within the computer. Additionally, trend prediction for any individual channel or monitor may be manually requested at any time via front panel trend and channel selection switches.
  • the system also incorporates frequency spectrum analysis circuitry which provides frequency spectrum sampling of input vibration signals over a wide range of frequencies in 1/20 octave steps. Only those frequencies whose amplitudes are greater than 10% of full scale are listed on the paper tape computer read-out, along with the overall vibration level. Vibration analysis is performed automatically upon receipt of a trip or alarm signal, for a calculated trend alarm for any channel, or at preset intervals. The paper tape read-out indicates which channel has gone into a fault condition and what that condition was (i.e., trip, alarm or trend alarm) as well as a change in any channel's condition.
  • each monitor such as monitor 10 is connected to a vibration pickup 14 in contact with a bearing of a rotating member 16, for example, and adapted to produce either a displacement, velocity or acceleration vibration signal.
  • Pickup 14 is connected through an amplifier 18 to a rectifier 20 which will produce an essentially steady-state direct current output signal on lead 22-1 which is applied to one input of a first multiplexer 24.
  • each of the other monitors will apply an input to the multiplexer 24, only the lead for the last monitor 48 being shown in the drawing and identified by the reference numeral 22-48.
  • the oscillatory vibration signal from the pickup 14 is also applied directly via lead 26-1 to a second multiplexer 28.
  • the oscillatory signal for the last monitor 12 being applied via lead 26-48 to multiplexer 28.
  • Each of the monitors also incorporates first and second comparators and relays 30 and 32.
  • comparator 30 for example, the direct current signal from rectifier 20, representing the amplitude of the vibration signal, is compared with a direct current signal from D.C. reference voltage source 34. If the direct current signal from rectifier 20 equals or exceeds the magnitude of the signal from source 34, then a relay is actuated to produce a steady-state direct current signal on lead 36-1 connected to the input of a third multiplexer 38.
  • the amplitude of the direct current signal from rectifier 20 at which the relay is closed to energize lead 36-1 is chosen arbitrarily and represents that amplitude of the vibration signal which signifies an alarm condition (i.e., an imminent malfunction).
  • the output of rectifier 20 is compared with a direct current signal from D.C. reference voltage source 40 in the comparator and relay 32, the arrangement being such that when the amplitude of the vibration signal reaches a point where the device being monitored should be shut down, the relay is actuated to energize lead 42-1.
  • This trip signal on lead 42-1 is also applied to the third multiplexer 38.
  • Alarm and trip signals are also applied to the multiplexer 38 from each of the other forty-seven monitors, the alarm signal from monitor 12 being on lead 36-48 and the trip signal from monitor 12 being on lead 42-48.
  • each monitor such as monitor 10
  • an external fault detector 44 adapted to detect faults such as a change in impedance due to breakage in the cable leading to the pickup 14 or an inaccurate gap for a non-contact vibration pickup such as that shown in US-A-3,707,671.
  • a signal is applied to the trip lead 42-49, common to all monitors, and applied to the multiplexer 38.
  • the occurrence of an external fault at any monitor causes a printer to print-out "System Alarm" without identifying the channel from which the fault signal was derived. This must be derived by manual examination of each monitor.
  • a manual programmer 46 comprising an internal jumper board, allows manual selection of individual channel parameters such as trip level setpoint for trend prediction and full-scale range for each channel, along with appropriate units of measure such as mm, cm/sec, m/sec 2 (mils, inches per second, G's).
  • a selection of sixteen combinations of (i.e., four binary bits) full-scale range in engineering units is provided for each channel. These sixteen choices, specified by the use of the monitoring system, are coded into the custom-programmed module or programmer 46 which forms part of the internal computer memory.
  • the jumber board allows individual channel selection to any one of sixteen choices.
  • functions common to all forty-eight channels may be selected on the jumber board 46, such as repetition rate of automatic data log print-out and "time until trip" setpoint of a trend alarm.
  • Each of the inputs from the programmer 46 passes through a digital multiplexer 48 to a computer 50 along with the inputs from multiplexers 38 and 24.
  • the multiplexer 48 is controlled from the computer 50 by means of a nine-bit address input 52. Similarly, multiplexer 38 is controlled so as to select a particular input channel monitor via a seven-bit address input 54.
  • Multiplexer 24 is controlled by a six-bit address input 56; however the output of the multiplexer 24 must pass through an analog-to-digital converter 58 before being fed into the digital computer 50 since the signals on leads 22-1 through 22-48 are direct current signals whose magnitudes are proportional to the magnitudes of the vibration signals being monitored.
  • the multiplexer 28, to which the oscillatory vibration signals on leads 26-1 through 26-48 are applied, is also controlled by a six-bit address input 60. A strobe input is applied to each of the multiplexers 24 and 28 via leads 62; while an end of conversion signal from each of the analog-to-digital converters 58 and 84 is fed back into the computer via leads 64.
  • the oscillatory vibration signals at the output of the multiplexer 28 are applied to the novel spectrum analyzing apparatus of the invention, enclosed by broken lines in Figure 1 and identified generally by the reference numeral 66. It comprises a single-double integrator 68 controlled by a signal from the computer 50. It is desired to perform a spectrum analysis on a vibration displacement signal. Hence, if the signal detected by any monitor is not a displacement signal but rather a velocity signal, a single integration is performed to convert it to a displacement signal. On the other hand, if the signal produced by a monitor is an acceleration signal, a double integration is performed to convert the acceleration signal to a displacement signal.
  • the output of the integrator 68 is coupled through the programmable gain amplifier 70 to the input of a voltage tuned filter 72 which has a passband which sweeps through the expected range of frequency components of an incoming vibration signal.
  • the operation of the voltage tuned filter is schematically illustrated in Figures 4 and 5.
  • the passband of the filter, indicated by the reference numeral 74 in Figure 4 is caused to sweep through a frequency range of 600 cycles per minute to 600,000 cycles per minute. This sweep takes a total of twenty-four seconds.
  • the dwell times are shown in Figure 5 and it will be noted that the dwell time for each frequency is 2 divided by the selected frequency.
  • the dwell time is about 1/5 of a second.
  • the dwell time for each successive step decreases until, at a frequency of 6000 cycles per minute, for example, it is 1/50th of a second.
  • the time to sweep through the band of frequencies from 600 to 6000 cycles per minute, as shown in Figure 4, is about eighteen seconds; however the time required to sweep through the band between 6000 and 60,000 cycles per minute is only four seconds; and the time to sweep through 60,000 cycles per minute to 600,000 cycles per minute is only about two seconds.
  • the manner in which the passband sweeps through the spectrum is controlled via address inputs or bits on lead 76 from the computer 50 applied to the voltage tuned filter 72 through a digital-to- analog converter 78. Signals passing through the voltage tuned filter are applied to the peak detector 80, the arrangement being such that only those frequencies whose amplitudes are greater then 10% of the full-scale value as determined by the internal computer program will be listed in the computer print-out.
  • the peak detector 80 is reset by a signal on lead 82 from the computer prior to each frequency sample derived from the voltage tuned filter 72. From the peak detector 80, the signal passes through the analog-to-digital converter 84 to the computer 50.
  • the computer 50 includes the usual input-output interface 86 connected to a central processing unit 88, the central processing unit 88 being controlled by a read-only memory comprising the computer program 90 and a random access memory 92.
  • the input-output interface is also connected to a printer 94.
  • touch switches 96 and 98 In addition to automatic functions, it is also possible to manually obtain data from any monitor or channel by means of touch switches 96 and 98.
  • the switches 96 and 98 have been adjusted to receive information from channel 17.
  • a system test can be achieved by depressing touch switch 100.
  • a data log can be achieved by depressing touch switch 102 and a spectrum analysis can be achieved by depressing switch 104.
  • a trend analysis can be achieved from any monitor by depressing touch switch 106, these switches being connected through a touch switch interface 108 to the computer 50.
  • a test voltage source 110 for example, will apply test voltages to two selected channels.
  • a flow diagram of the computer program utilized with the invention is as follows:
  • the first step in the program is to declare all variables to be read into the random access memory 92 and their location in storage. This includes direct current amplitude signals from multiplexer 24, the signals from manual programmer 46, and the trip and alarm signals from multiplexer 38.
  • a table of frequencies to be printed out in each spectrum analysis is then constructed from data permanently stored in the read-only memory 90. This table is the same for all channels; however only those frequencies will be printed out which exceed 10% of full scale in amplitude.
  • the next step in the program is to construct a table of tuning voltages derived from the read-only memory 90 for the voltage tuned filter 72, this corresponding to the table of frequencies to be printed out.
  • Direct current multiplexing by multiplexer 24 is then activated; whereupon each of the direct current amplitude signals from the multiplexer 24 is sampled in succession. This is followed by a reading of the internal clock in hours and days, the days being calculated from accumulated hours.
  • the internal clock is capable of indicating the day of the year from 1 through 365 as well as time of day up to 24 hours.
  • the following step in the program is to select a channel for manual frequency analysis or trend analysis.
  • the central processing unit 88 activated by touch switches 96 and 98, is conditioned to receive signals from a single channel to perform a spectrum analysis upon depression of touch switch 104 or a trend print-out upon depression of touch switch 106.
  • a test alarm status is performed by momentarily altering internal test voltages.
  • the print-out will indicate system alarm and system normal as test voltages are altered, then returned to normal. This step insures that the internal computer circuitry is operating properly.
  • the digital multiplexers 38 and 48 are then activated to read-in alarm and trip signals as well as information from the manual programmer 46.
  • a status file is then activated to store normal, alarm and trip signals and to determine whether there has been a change in an alarm, trip or normal signal. Following this, the trend alarm is established, which is the time to failure (i.e., trip) of a particular unit being monitored. Generally, this time will be the same for all channels.
  • the next step in the program is to read in full-scale units for each monitor and the engineering units from the manual programmer 46. This determines: (1) the time period between scheduled automatic data log print-outs (i.e., one hour, eight hours, etc.); (2) data log print-out upon receipt of a trip, alarm or trend alarm signal; and (3) automatic spectrum analysis print-out upon receipt of a trend alarm, a trip signal, or an alarm signal. A scaling factor for full scale is then entered which corrects the stored overall value for full-scale readings. This is followed by the manual data log, manual trend and manual analysis input commands. At this time, the conditions of switches 100-106 are examined by the central processing unit 88 to determine if a manually-activated print-out has been commanded. The alarm trend for all channels is then computed and stored with the last four hourly-readings of vibration level from multiplexer 24.
  • Figures 2 and 3 illustrate the manner in which the trand alarm is calculated. From Figure 2, it can be seen that the vibration amplitude from a particular monitor has risen over five successive hours. At the 6th hour, the signal received at the first hour is removed from storage and the 6th-hour signal is inserted. However, before the first-hour signal is removed, it is averaged with the first through fifth-hour signals. Likewise, the second through sixth-hour signals are averaged. From these two averages, the computer establishes, in effect, a straight line 112 and calculates the slope of that line. Whether or not an alarm trend signal will be generated is achieved by calculating, through a simple trigonometric relationship, the time between the last average point and an intersection of line 112 with an established trip set-point 114.
  • the calculated time is equal to or less than a predetermined time stored in the random access memory 92 (which is the same for all channels), then automatic input-output occurs for the channel in question as well as a vibration analysis for that channel and a data log on all monitors associated with a piece of equipment from which the trend alarm was signaled.
  • the final steps in the program comprise analysis print-out, data log print-out, trend alarm print-out and system alarm print-out, in which steps the printer is commanded to print-out data stored in the random access memory 92.
  • Typical print-outs from the printer 94 under certain conditions are as follows:
  • the first print-out above is normal periodic data log or a data log which can be on command via the touch switch 102.
  • the number 1017 indicates that the print-out occurred at the 10th hour and 17th minute of the day in question; and the number 025 indicates that the print-out occurred on the 25th day of the year.
  • the condition of each channel is printed out beneath the date and time. For example, channel No. 1 prints out 0.15 G's, corresponding to 147.2 m/sec 2 .
  • the arithmetic unit involved for this particular channel was determined by the manual programmer 46 as are the arithmetic units for all of the other channels.
  • Channel No. 3 for example, prints out 0.81 MILS corresponding to ca. 0.02 mm, whereas channel No. 4 prints out 0.07 inch per second corresponding to 1.72 mm/sec and represents a signal derived from an accelerometer pickup.
  • the next print-out represents a spectrum analysis for a particular channel on command via the touch switch 104 of Fig. 1.
  • the print-out shows that the analysis occurred at the 20th hour and 31 st minute of the 90th day of the year and is for channel No. 21, this being determined by the touch switches 96 and 98 in Figure 1.
  • the print-out shows that the overall signal level (i.e., for all frequencies) is 0.81 Mils, corresponding to ca. 0.02 mm.
  • a print-out of the specific amplitudes at various predetermined frequencies which are initially determined in the manual programmer 46. In the example shown, samples are taken at 1476, 1582, 1696, etc. cycles per minute. From this analysis, and from previous experience with the vibrating equipment in question, the general condition of the equipment can be determined. For example, excessive amplitude at one frequency can indicate a lubrication problem. The tips of the dashed lines to the right of the amplitude readings give an approximate visual representation or plot of the spectral response of the input signal.
  • Each dash represents a full 0.04 mil or approximately 0.001 mm amplitude such that the line for 0.43 mils or approximately 0.011 mm, for example, contains 10 dashes, that for 0.39 mils or approximately 0.0098 mm contains 9 dashes, etc.
  • the next two print-outs in the foregoing example are trend on command via the touch switch 106 of Figure 1 and an automatic trend alarm.
  • the print-out indicates that for channel 15, preselected via the switches 96 and 98, there are an infinite number of hours to trip at 10:12 A.M. on the 95th day of the year and that the equipment being monitored is operating satisfactorily.
  • the next print-out is an automatic vibration analysis and data log upon receipt of an alarm or trip signal from any monitor. This automatic analysis occurred on the 31 Oth day of the year at 1:07 A.M. for channel 11. Following the print-out of the vibration analysis at preselected frequencies is a data log for only those monitors associated with the equipment from which the alarm or trip signal was received on channel 11.
  • monitors 3, 4, 11, 24 and 25 preselected in the manual programmer 46.
  • the "T” for channel 11 shows that this channel went into a trip condition and the "A” for channel 3 shows that this channel went into an alarm condition.
  • the "TD” signifies that both channels 3 and 11 are in a trend alarm condition also.
  • the asterisk indicates a change in that channel's condition. When the fault condition is reset, an automatic data log will follow, with only the asterisk present (i.e., without the "T", "A” or "TD” designations).
  • a system test print-out occurs when touch switch 100 is depressed.
  • the system test provides for checking of internal circuit faults sensing by momentarily altering the internal test voltages via the touch switch 100.
  • the print-out indicates system alarm and system normal as test voltages are altered, then returned to normal.
  • An automatic system alarm occurs when an external monitor system circuit fault relay is energized while a system normal will result when the external relay is released. Also, an automatic system alarm occurs if a malfunction in the data acquisition system is detected. A system normal will result when the malfunction is corrected.

Description

  • While not limited thereto, the present invention is particularly adapted for use in monitoring vibrations produced by rotating or other types of machinery in a complete industrial installation, such as refinery. By monitoring vibrations in this manner, malfunctions and probable future failures of any machines within the industrial installation can be readily ascertained; and corrective action can be taken immediately and before a breakdown or possible dangerous condition occurs.
  • There are at present essentially two types of such controlling systems-the dedicated minicomputer system and the simple data logger. Computer systems generally include disc memory for data storage, CRT terminals for display of data and line printers for hard copy of data. As a result, they require a relatively large capital investment. While simple data loggers are relatively inexpensive, they offer simple functions only such as logging data and comparing the data to setpoints.
  • From FR-A-2 312 060 a system has become known for monitoring the operation of an apparatus using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to apply respective electrical signals via multiplexing means to a computer apparatus, having memory means and read-out means for periodically storing at least selected ones of the electrical signals from each monitoring device in the memory means, the computer apparatus further including means for performing a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times in said memory means an alarm being initiated indicative of a significant trend. The known system is exclusively based on a monitoring by way of a power spectral density analysis (PSD) which is a squared form of analysis which is generally non-linear. A trend analysis is performed by taking samples of each of certain frequency intervals of the PSD plot and averaging the data over that particular interval for any one scan cycle to obtain an average value over that interval which is stored as historical data. The historical data are then used for the trend analysis by taking the average value of the sample in any one frequency interval and doing a regression analysis to fit the best trend line of those data. A trend line showing a significant slope in the way of increasing or decreasing value over a period of time indicates a significant trend and would alert the operator. This system is one of the complex computer type systems. Nevertheless, only certain frequency intervals from the whole frequency range are monitored. With the known system it is necessary to store the average values for all frequency intervals separately. Though the operator is alerted and the appropriate corrective action is indicated, no information is given in respect of the urgency of such an action. Thus, if there are several alarm conditions at a time, he does not know the priority of the actions necessary.
  • In the article "Computers Monitors Gas Turbine Performance", in Instruments and Control Systems, Vol. 47, No. 8, August 1974, pages 30-31, a system is described in general terms using trend analysis for monitoring the maintenance of industrial power generation equipment. This is again a complex computer type system. The system consists of a set of sensors, a data collection unit and a central processing unit. The latter accepts the data from the data collection unit and utilises a number of different techniques such as gas path analysis and limit comparisons against truth tables to constantly evaluate the health of the engine. A method of regression analysis detects any trends in the plant towards the development of specific defects or the exceeding of any limits. Any malfunctions are signalled by warning messages and print-outs of relevant data.
  • It is the object of the present invention to provide a system and a method for monitoring the operation of an apparatus which are simple in design, consider the total frequency range and offer optimum information to the operator in respect of any malfunction.
  • According to the invention, the system for monitoring the operation of an apparatus, using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to apply respective electrical signals via multiplexing means to a computer apparatus having read-out means and memory means for periodically storing at least selected ones of the electrical signals from each monitoring device, the computer apparatus further including means for performing a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times in said memory means, an alarm being initiated indicative of a significant trend, is characterised in that for a said monitoring device there is provided a means for generating said selected electrical signal corresponding to the overall amplitude of a plurality of signal amplitudes substantially corresponding to the frequency response range of a said monitoring device, that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that the said computer apparatus includes means for computing from the trend of said values the probable time to failure of the monitored apparatus and means for causing said read-out to provide information indicative of the probable time to failure.
  • According to the invention, a method for monitoring the operation of an apparatus using a plurality of monitoring devices each adapted to monitor a physical condition of the apparatus and to produce respective electrical signals, including the steps of feeding each of said electrical signals from the respective monitoring devices to a computer apparatus, periodically storing at least selected ones of the electrical signals from each monitoring device performing in the computer apparatus a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times and initiating an alarm for a significant trend, is characterised in that said selected electrical signals correspond to the overall amplitude of a plurality of signal amplitudes substantially over the frequency response range of a said monitoring device, that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that from the trend of the said values the probable time to failure of the monitored apparatus is computed and that information indicative of the probable time to failure is provided.
  • In accordance with the present invention, a data monitoring system is provided which does not require a large capital investment but which, nevertheless, is capable of reading out complete system information including a malfunction of any one of a number of different devices being monitored, the time to failure of any pieces of equipment being monitored, and an analysis of the input information. In the case where the invention is used in a vibration monitoring system, it performs the functions of automatic channel data logging, frequency spectrum analysis, and vibration level trend prediction. Each of these functions additionally may be manually selected for each individual monitor or channel via front panel control. A built-in system fault detection circuit is used which will respond to either an internal or circuit fault or to an external system alarm relay closure. Data readout is obtained via a self- contained dot-matrix printer assembly.
  • All functions of the monitoring system of the invention are under the control of an internal microcomputer which continuously samples data from a plurality of monitors. At each monitor, vibration input signals are obtained directly from velocity pickups, self-amplified accelerometers, non-contact signal sensors or from accelerometer preamps. In addition, direct current signals proportional to vibration level or amplitude and trip alarm signals are obtained from the monitors, these latter signals being derived by comparison of the actual vibration signal with reference signals proportional to preselected alarm and trip levels.
  • The system automatically indicates, via the computer read-out those channels which go into a trip condition within a preselected time span. That is, the time to failure is calculated and displayed via the print-out. Each channel's "look ahead" time may be selected with a user-programmable jumber board within the computer. Additionally, trend prediction for any individual channel or monitor may be manually requested at any time via front panel trend and channel selection switches.
  • The system also incorporates frequency spectrum analysis circuitry which provides frequency spectrum sampling of input vibration signals over a wide range of frequencies in 1/20 octave steps. Only those frequencies whose amplitudes are greater than 10% of full scale are listed on the paper tape computer read-out, along with the overall vibration level. Vibration analysis is performed automatically upon receipt of a trip or alarm signal, for a calculated trend alarm for any channel, or at preset intervals. The paper tape read-out indicates which channel has gone into a fault condition and what that condition was (i.e., trip, alarm or trend alarm) as well as a change in any channel's condition.
  • The above and other objects and features of the invention will become apparent from the following detailed description taken in connection with the accompanying drawings which form a part of this specification, and in which:
    • Figures 1 A and 1 B (hereinafter referred to together as "Fig. 1") comprise a schematic block diagram of the system of the invention;
    • Figures 2 and 3 graphically illustrate the manner in which successive sampled vibration level signals are stored in the computer of the system and the manner in which a trend (i.e., time to failure) is determined; and
    • Figures 4 and 5 graphically illustrate the operation of the voltage tuned filter utilized in the spectrum analyzer of the invention.
  • With reference now to the drawings, and particularly to Figure 1, the monitoring system shown includes forty-eight channels or monitors for monitoring a physical condition of a device to be monitored. Only monitor Nos. 1 and 48 are shown in the drawing and are identified by the reference numerals 10 and 12. It will be further assumed for purposes of explanation that the data acquisition system is to be used in a vibration monitoring system. Thus, each monitor, such as monitor 10, is connected to a vibration pickup 14 in contact with a bearing of a rotating member 16, for example, and adapted to produce either a displacement, velocity or acceleration vibration signal. Pickup 14 is connected through an amplifier 18 to a rectifier 20 which will produce an essentially steady-state direct current output signal on lead 22-1 which is applied to one input of a first multiplexer 24. Similarly, each of the other monitors will apply an input to the multiplexer 24, only the lead for the last monitor 48 being shown in the drawing and identified by the reference numeral 22-48.
  • The oscillatory vibration signal from the pickup 14 is also applied directly via lead 26-1 to a second multiplexer 28. The same is true of the remaining monitors, the oscillatory signal for the last monitor 12 being applied via lead 26-48 to multiplexer 28. Each of the monitors also incorporates first and second comparators and relays 30 and 32. In comparator 30, for example, the direct current signal from rectifier 20, representing the amplitude of the vibration signal, is compared with a direct current signal from D.C. reference voltage source 34. If the direct current signal from rectifier 20 equals or exceeds the magnitude of the signal from source 34, then a relay is actuated to produce a steady-state direct current signal on lead 36-1 connected to the input of a third multiplexer 38. The amplitude of the direct current signal from rectifier 20 at which the relay is closed to energize lead 36-1 is chosen arbitrarily and represents that amplitude of the vibration signal which signifies an alarm condition (i.e., an imminent malfunction). Similarly, the output of rectifier 20 is compared with a direct current signal from D.C. reference voltage source 40 in the comparator and relay 32, the arrangement being such that when the amplitude of the vibration signal reaches a point where the device being monitored should be shut down, the relay is actuated to energize lead 42-1. This trip signal on lead 42-1 is also applied to the third multiplexer 38. Even though the equipment in question may be shut down automatically upon receipt of a trip signal, ordinarily sufficient momentum of the rotating parts, for example, will keep the parts rotating for a sufficient period of time to permit a meaningful spectrum analysis and data log to be taken. Alarm and trip signals are also applied to the multiplexer 38 from each of the other forty-seven monitors, the alarm signal from monitor 12 being on lead 36-48 and the trip signal from monitor 12 being on lead 42-48.
  • Included in each monitor, such as monitor 10, is an external fault detector 44 adapted to detect faults such as a change in impedance due to breakage in the cable leading to the pickup 14 or an inaccurate gap for a non-contact vibration pickup such as that shown in US-A-3,707,671. Whenever an external fault occurs, a signal is applied to the trip lead 42-49, common to all monitors, and applied to the multiplexer 38. As will be seen, in the particular embodiment of the invention shown herein, the occurrence of an external fault at any monitor causes a printer to print-out "System Alarm" without identifying the channel from which the fault signal was derived. This must be derived by manual examination of each monitor.
  • A manual programmer 46, comprising an internal jumper board, allows manual selection of individual channel parameters such as trip level setpoint for trend prediction and full-scale range for each channel, along with appropriate units of measure such as mm, cm/sec, m/sec2 (mils, inches per second, G's). A selection of sixteen combinations of (i.e., four binary bits) full-scale range in engineering units is provided for each channel. These sixteen choices, specified by the use of the monitoring system, are coded into the custom-programmed module or programmer 46 which forms part of the internal computer memory. The jumber board allows individual channel selection to any one of sixteen choices. In addition, functions common to all forty-eight channels may be selected on the jumber board 46, such as repetition rate of automatic data log print-out and "time until trip" setpoint of a trend alarm. Each of the inputs from the programmer 46 passes through a digital multiplexer 48 to a computer 50 along with the inputs from multiplexers 38 and 24.
  • The multiplexer 48 is controlled from the computer 50 by means of a nine-bit address input 52. Similarly, multiplexer 38 is controlled so as to select a particular input channel monitor via a seven-bit address input 54. Multiplexer 24 is controlled by a six-bit address input 56; however the output of the multiplexer 24 must pass through an analog-to-digital converter 58 before being fed into the digital computer 50 since the signals on leads 22-1 through 22-48 are direct current signals whose magnitudes are proportional to the magnitudes of the vibration signals being monitored. The multiplexer 28, to which the oscillatory vibration signals on leads 26-1 through 26-48 are applied, is also controlled by a six-bit address input 60. A strobe input is applied to each of the multiplexers 24 and 28 via leads 62; while an end of conversion signal from each of the analog-to- digital converters 58 and 84 is fed back into the computer via leads 64.
  • The oscillatory vibration signals at the output of the multiplexer 28 are applied to the novel spectrum analyzing apparatus of the invention, enclosed by broken lines in Figure 1 and identified generally by the reference numeral 66. It comprises a single-double integrator 68 controlled by a signal from the computer 50. It is desired to perform a spectrum analysis on a vibration displacement signal. Hence, if the signal detected by any monitor is not a displacement signal but rather a velocity signal, a single integration is performed to convert it to a displacement signal. On the other hand, if the signal produced by a monitor is an acceleration signal, a double integration is performed to convert the acceleration signal to a displacement signal.
  • From the sixteen combinations selected by the manual programmer 46, it is known whether or not integration is required and the gain required for amplifier 70. For example, if channel No. 21 is programmed in mm or mils (i.e., displacement), a single integration is required to convert a velocity signal in inches per second to mils. Additionally, the gain of amplifier 70 is adjusted to give a full-scale output for the particular vibration pickup used. For example, if a velocity pick-up for channel No. 10 has an output of 764 millivolts RMS (Root-mean-square) per 24.5 mm/sec (per inch per second peak) then the amplifier gain must be ten to achieve a 7.64 volt full-scale output required for a peak detector 80 adapted to detect a peak voltage of 10 volts, as dictated by an analog-to-digital converter 84.
  • The output of the integrator 68 is coupled through the programmable gain amplifier 70 to the input of a voltage tuned filter 72 which has a passband which sweeps through the expected range of frequency components of an incoming vibration signal. The operation of the voltage tuned filter is schematically illustrated in Figures 4 and 5. The passband of the filter, indicated by the reference numeral 74 in Figure 4 is caused to sweep through a frequency range of 600 cycles per minute to 600,000 cycles per minute. This sweep takes a total of twenty-four seconds. However, in order to obtain a good frequency sample, it is necessary to have the passband dwell at each frequency being sampled for at least 2 cycles of the selected frequency. The dwell times are shown in Figure 5 and it will be noted that the dwell time for each frequency is 2 divided by the selected frequency. Thus, at the lowest frequency of 600 cycles per minute (10 cps), the dwell time is about 1/5 of a second. The dwell time for each successive step decreases until, at a frequency of 6000 cycles per minute, for example, it is 1/50th of a second. The time to sweep through the band of frequencies from 600 to 6000 cycles per minute, as shown in Figure 4, is about eighteen seconds; however the time required to sweep through the band between 6000 and 60,000 cycles per minute is only four seconds; and the time to sweep through 60,000 cycles per minute to 600,000 cycles per minute is only about two seconds.
  • The manner in which the passband sweeps through the spectrum is controlled via address inputs or bits on lead 76 from the computer 50 applied to the voltage tuned filter 72 through a digital-to- analog converter 78. Signals passing through the voltage tuned filter are applied to the peak detector 80, the arrangement being such that only those frequencies whose amplitudes are greater then 10% of the full-scale value as determined by the internal computer program will be listed in the computer print-out. The peak detector 80 is reset by a signal on lead 82 from the computer prior to each frequency sample derived from the voltage tuned filter 72. From the peak detector 80, the signal passes through the analog-to-digital converter 84 to the computer 50. The computer 50 includes the usual input-output interface 86 connected to a central processing unit 88, the central processing unit 88 being controlled by a read-only memory comprising the computer program 90 and a random access memory 92. The input-output interface is also connected to a printer 94.
  • In addition to automatic functions, it is also possible to manually obtain data from any monitor or channel by means of touch switches 96 and 98. In the illustration given in Figure 1, for example, the switches 96 and 98 have been adjusted to receive information from channel 17. After the channel is selected, a system test can be achieved by depressing touch switch 100. Similarly, a data log can be achieved by depressing touch switch 102 and a spectrum analysis can be achieved by depressing switch 104. Finally, a trend analysis can be achieved from any monitor by depressing touch switch 106, these switches being connected through a touch switch interface 108 to the computer 50. When touch switch 100 is depressed, a test voltage source 110, for example, will apply test voltages to two selected channels.
  • A flow diagram of the computer program utilized with the invention is as follows:
    Figure imgb0001
    Figure imgb0002
  • The first step in the program is to declare all variables to be read into the random access memory 92 and their location in storage. This includes direct current amplitude signals from multiplexer 24, the signals from manual programmer 46, and the trip and alarm signals from multiplexer 38. A table of frequencies to be printed out in each spectrum analysis is then constructed from data permanently stored in the read-only memory 90. This table is the same for all channels; however only those frequencies will be printed out which exceed 10% of full scale in amplitude. The next step in the program is to construct a table of tuning voltages derived from the read-only memory 90 for the voltage tuned filter 72, this corresponding to the table of frequencies to be printed out. Direct current multiplexing by multiplexer 24 is then activated; whereupon each of the direct current amplitude signals from the multiplexer 24 is sampled in succession. This is followed by a reading of the internal clock in hours and days, the days being calculated from accumulated hours. The internal clock is capable of indicating the day of the year from 1 through 365 as well as time of day up to 24 hours.
  • The following step in the program is to select a channel for manual frequency analysis or trend analysis. In this phase, the central processing unit 88, activated by touch switches 96 and 98, is conditioned to receive signals from a single channel to perform a spectrum analysis upon depression of touch switch 104 or a trend print-out upon depression of touch switch 106. Thereafter, a test alarm status is performed by momentarily altering internal test voltages. The print-out will indicate system alarm and system normal as test voltages are altered, then returned to normal. This step insures that the internal computer circuitry is operating properly. The digital multiplexers 38 and 48 are then activated to read-in alarm and trip signals as well as information from the manual programmer 46. A status file is then activated to store normal, alarm and trip signals and to determine whether there has been a change in an alarm, trip or normal signal. Following this, the trend alarm is established, which is the time to failure (i.e., trip) of a particular unit being monitored. Generally, this time will be the same for all channels.
  • The next step in the program is to read in full-scale units for each monitor and the engineering units from the manual programmer 46. This determines: (1) the time period between scheduled automatic data log print-outs (i.e., one hour, eight hours, etc.); (2) data log print-out upon receipt of a trip, alarm or trend alarm signal; and (3) automatic spectrum analysis print-out upon receipt of a trend alarm, a trip signal, or an alarm signal. A scaling factor for full scale is then entered which corrects the stored overall value for full-scale readings. This is followed by the manual data log, manual trend and manual analysis input commands. At this time, the conditions of switches 100-106 are examined by the central processing unit 88 to determine if a manually-activated print-out has been commanded. The alarm trend for all channels is then computed and stored with the last four hourly-readings of vibration level from multiplexer 24.
  • Figures 2 and 3 illustrate the manner in which the trand alarm is calculated. From Figure 2, it can be seen that the vibration amplitude from a particular monitor has risen over five successive hours. At the 6th hour, the signal received at the first hour is removed from storage and the 6th-hour signal is inserted. However, before the first-hour signal is removed, it is averaged with the first through fifth-hour signals. Likewise, the second through sixth-hour signals are averaged. From these two averages, the computer establishes, in effect, a straight line 112 and calculates the slope of that line. Whether or not an alarm trend signal will be generated is achieved by calculating, through a simple trigonometric relationship, the time between the last average point and an intersection of line 112 with an established trip set-point 114. If the calculated time is equal to or less than a predetermined time stored in the random access memory 92 (which is the same for all channels), then automatic input-output occurs for the channel in question as well as a vibration analysis for that channel and a data log on all monitors associated with a piece of equipment from which the trend alarm was signaled. The final steps in the program comprise analysis print-out, data log print-out, trend alarm print-out and system alarm print-out, in which steps the printer is commanded to print-out data stored in the random access memory 92.
  • Typical print-outs from the printer 94 under certain conditions are as follows:
    Figure imgb0003
    Figure imgb0004
    Figure imgb0005
    Figure imgb0006
  • The first print-out above is normal periodic data log or a data log which can be on command via the touch switch 102. The number 1017 indicates that the print-out occurred at the 10th hour and 17th minute of the day in question; and the number 025 indicates that the print-out occurred on the 25th day of the year. The condition of each channel is printed out beneath the date and time. For example, channel No. 1 prints out 0.15 G's, corresponding to 147.2 m/sec2. The arithmetic unit involved for this particular channel was determined by the manual programmer 46 as are the arithmetic units for all of the other channels. Channel No. 3, for example, prints out 0.81 MILS corresponding to ca. 0.02 mm, whereas channel No. 4 prints out 0.07 inch per second corresponding to 1.72 mm/sec and represents a signal derived from an accelerometer pickup.
  • The next print-out represents a spectrum analysis for a particular channel on command via the touch switch 104 of Fig. 1. The print-out shows that the analysis occurred at the 20th hour and 31 st minute of the 90th day of the year and is for channel No. 21, this being determined by the touch switches 96 and 98 in Figure 1.
  • The print-out shows that the overall signal level (i.e., for all frequencies) is 0.81 Mils, corresponding to ca. 0.02 mm. Following this is a print-out of the specific amplitudes at various predetermined frequencies which are initially determined in the manual programmer 46. In the example shown, samples are taken at 1476, 1582, 1696, etc. cycles per minute. From this analysis, and from previous experience with the vibrating equipment in question, the general condition of the equipment can be determined. For example, excessive amplitude at one frequency can indicate a lubrication problem. The tips of the dashed lines to the right of the amplitude readings give an approximate visual representation or plot of the spectral response of the input signal. Each dash represents a full 0.04 mil or approximately 0.001 mm amplitude such that the line for 0.43 mils or approximately 0.011 mm, for example, contains 10 dashes, that for 0.39 mils or approximately 0.0098 mm contains 9 dashes, etc.
  • The next two print-outs in the foregoing example are trend on command via the touch switch 106 of Figure 1 and an automatic trend alarm. In the trend on command, the print-out indicates that for channel 15, preselected via the switches 96 and 98, there are an infinite number of hours to trip at 10:12 A.M. on the 95th day of the year and that the equipment being monitored is operating satisfactorily. The next print-out is an automatic vibration analysis and data log upon receipt of an alarm or trip signal from any monitor. This automatic analysis occurred on the 31 Oth day of the year at 1:07 A.M. for channel 11. Following the print-out of the vibration analysis at preselected frequencies is a data log for only those monitors associated with the equipment from which the alarm or trip signal was received on channel 11. These comprise monitors 3, 4, 11, 24 and 25 preselected in the manual programmer 46. The "T" for channel 11 shows that this channel went into a trip condition and the "A" for channel 3 shows that this channel went into an alarm condition. The "TD" signifies that both channels 3 and 11 are in a trend alarm condition also. The asterisk indicates a change in that channel's condition. When the fault condition is reset, an automatic data log will follow, with only the asterisk present (i.e., without the "T", "A" or "TD" designations).
  • Finally, a system test print-out occurs when touch switch 100 is depressed. As was explained above, the system test provides for checking of internal circuit faults sensing by momentarily altering the internal test voltages via the touch switch 100. The print-out indicates system alarm and system normal as test voltages are altered, then returned to normal. An automatic system alarm occurs when an external monitor system circuit fault relay is energized while a system normal will result when the external relay is released. Also, an automatic system alarm occurs if a malfunction in the data acquisition system is detected. A system normal will result when the malfunction is corrected.

Claims (16)

1. System for monitoring the operation of an apparatus, using a plurality of monitoring devices (10, 12) each adapted to monitor a physical condition of the apparatus and to apply respective electrical signals via multiplexing means (24) to a computer apparatus (50) having read-out means (94) and memory means (92) for periodically storing at least selected ones of the electrical signals from each monitoring device (10), the computer apparatus (50) further including means (88) for performing a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times in said memory means (92), an alarm being initiated indicative of a significant trend, characterised in that for a said monitoring device (10, 12) there is provided a means (20) for generating said selected electrical signal corresponding to the overall amplitude of a plurality of signal amplitudes substantially corresponding to the frequency response range of a said monitoring device (10), that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that the said computer apparatus (50) includes means for computing from the trend of said values the probable time to failure of the monitored apparatus and means for causing said read-out to provide information indicative of the probable time to failure.
2. System according to claim 1, characterised in that the oldest stored said selected signal is removed from the memory means (92) each time a new selected signal is fed into said memory means.
3. System according to claim 2 characterized in that for the trend analysis there are provided means for averaging the values of the said selected signals stored at a fixed number of monitoring times, each time before and after a new selected signal is fed into and the oldest selected signal is removed from said memory means (92), and that the probable time to failure of the apparatus being monitored is calculated as the difference of the instant time and the time corresponding to the intersection point of a line containing the two average values and a line determined by a pre-set upper limit value.
4. System according to any one of the claims 1 to 3, characterized by means (30) for sensing an alarm condition of a said selected signal for each monitoring device (10), indicating a probable malfunction of the apparatus being monitored and for producing a steady-state alarm signal, means (32) for sensing a trip condition of a said selected signal from each monitoring device (10) and for producing a steady-state trip signal indicating that the apparatus being monitored should be shut down, multiplexing means (38) for feeding all of said alarm and trip condition signals to the said computer apparatus and means within said computer apparatus (50) for actuating the read-out means, as print-out means (94) to print the existence of an alarm or trip condition and an indication of the monitoring device (10) from which it was derived.
5. System according to any one of the preceding claims, characterised in that said read-out means, as print-out means (94) are arranged to be manually actuated for printing the magnitude of the said electrical signals from any monitoring device (10) and/or the trend in the magnitude of the overall amplitude signals from any monitoring device (10).
6. System according to any one of the claims 1 to 4, characterized in that said selected electrical signals indicative of a physical condition are derived from vibration signals and that there is provided a means (66) for performing a spectrum analysis on a vibration signal from a monitoring device (10).
7. System according to claim 6 characterised by means for actuating said read-out means, as print-out means (94) to print selected ones of the frequencies in said vibration signal and the amplitude of said selected frequencies.
8. System according to claim 6 or 7 characterized in that the means (66) for performing a spectrum analysis includes a voltage tuned filter (72) which samples selected ones of the frequencies in the vibration signal.
9. System according to claim 6, 7 or 8, characterized in that a spectrum analysis is performed automatically on a vibration signal from a monitoring device (10) only when that monitoring device (10) generates an alarm or trip condition signal.
10. System according to any one of the preceding claims, characterised in that said read-out means, as print-out means (94) are periodically actuated to automatically print the said stored values of the said electrical signals for each monitoring device (10).
11. System according to any one of the claims 6 to 10, characterized in that there is provided a filter (72) having a variable pass band and capable of sampling the said vibration signal over a frequency range corresponding with that of said means for producing the said electrical signals derived from vibration signals, means for causing said pass band to sweep through said frequency range in steps to produce sample signals at different frequencies, the pass band stopping during each step in an amount at least equal to two divided by the frequency being sampled, a peak detector coupled to the output of said filter for generating output signals only when the sampled frequencies exceed a predetermined amplitude and means for recording those frequencies detected by the peak detector (80).
12. System according to claim 11 characterized by means for recording the maximum amplitude of the electrical signals derived from the vibration signals for each recorded frequency.
13. System according to any one of the claims 6 to 12 characterized in that the spectrum analysing apparatus includes means (68) for performing a single or a double integration on an incoming electrical signal from the monitoring device (10) depending on whether an incoming signal is a velocity or an acceleration signal.
14. System according to any one of the preceding claims, characterized by means for momentarily altering internal test voltages thereby checking for internal circuit faults.
15. System according to any of the preceding claims, characterized by a means initiating an automatic system alarm upon detection of an external system circuit fault.
16. Method for monitoring the operation of an apparatus using a plurality of monitoring devices (10, 12) each adapted to monitor a physical condition of the apparatus and to produce respective electrical signals, including the steps of feeding each of said electrical signals from the respective monitoring devices (10) to a computer apparatus, periodically storing at least selected ones of the electrical signals from each monitoring device (10), performing in the computer apparatus (50) a trend analysis on the basis of historical data consisting of selected electrical signals stored before at various times and initiating an alarm for a significant trend, characterised in that said selected electrical signals correspond to the overall amplitude of a plurality of signal amplitudes substantially over the frequency response range of a said monitoring device (10), that the trend analysis is performed on the basis of a predetermined restricted number of said stored values, that from the trend of the said values the probable time to failure of the monitored apparatus is computed and that information indicative of the probable time to failure is provided.
EP78101434A 1977-11-25 1978-11-22 System and method for monitoring the operation of an apparatus Expired EP0002232B1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US05/854,939 US4184205A (en) 1977-11-25 1977-11-25 Data acquisition system
US854939 1992-03-19

Publications (3)

Publication Number Publication Date
EP0002232A2 EP0002232A2 (en) 1979-06-13
EP0002232A3 EP0002232A3 (en) 1980-01-23
EP0002232B1 true EP0002232B1 (en) 1983-06-22

Family

ID=25319934

Family Applications (1)

Application Number Title Priority Date Filing Date
EP78101434A Expired EP0002232B1 (en) 1977-11-25 1978-11-22 System and method for monitoring the operation of an apparatus

Country Status (5)

Country Link
US (1) US4184205A (en)
EP (1) EP0002232B1 (en)
JP (1) JPS5494061A (en)
CA (1) CA1139881A (en)
DE (1) DE2862285D1 (en)

Families Citing this family (77)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4285241A (en) * 1979-07-13 1981-08-25 Westinghouse Electric Corp. Method and apparatus for the determination of the mass of an impacting object
JPS56130634A (en) 1980-03-19 1981-10-13 Hitachi Ltd Method and device for monitoring oscillation of rotary machine
US4453407A (en) * 1980-04-17 1984-06-12 Hitachi, Ltd. Vibration diagnosis method and apparatus for rotary machines
DE3022371A1 (en) * 1980-06-14 1981-12-24 Philips Patentverwaltung Gmbh, 2000 Hamburg DATA INPUT OR OUTPUT DEVICE WITH FUNCTIONAL CHECK
US4349881A (en) * 1980-07-14 1982-09-14 International Telephone And Telegraph Corporation Vibration instruments
US4733361A (en) * 1980-09-03 1988-03-22 Krieser Uri R Life usage indicator
GB2088105B (en) * 1980-11-17 1984-09-19 British Gas Corp Data recording apparatus
US4408285A (en) * 1981-02-02 1983-10-04 Ird Mechanalysis, Inc. Vibration analyzing apparatus and method
US4380172A (en) * 1981-02-19 1983-04-19 General Electric Company On-line rotor crack detection
JPS57155696A (en) * 1981-03-23 1982-09-25 Hitachi Ltd Data processor for automobile
US4408294A (en) * 1981-03-27 1983-10-04 General Electric Company Method for on-line detection of incipient cracks in turbine-generator rotors
GB2103801B (en) * 1981-08-04 1985-05-22 British Gas Corp Assessing lifetime of duct by measuring fluid pressure and temperature within the duct
US4409824A (en) * 1981-09-14 1983-10-18 Conoco Inc. Fatigue gauge for drill pipe string
US4476531B1 (en) * 1981-12-04 1999-01-09 Spx Corp Engine analyzer with digital waveform display
US4455634A (en) * 1982-01-12 1984-06-19 Discovision Associates Audio/video quality monitoring system
US4524444A (en) * 1982-01-12 1985-06-18 Discovision Associates Analyzing the signal transfer characteristics of a signal processing unit
DE3374723D1 (en) * 1982-02-17 1988-01-07 British Aerospace Aircraft data instrumentation and acquisition system
JPS58193914A (en) * 1982-04-09 1983-11-11 Hitachi Constr Mach Co Ltd Trouble preindicating device for hydraulic system
US4591967A (en) * 1982-06-29 1986-05-27 Andover Controls Corporation Distributed drum emulating programmable controller system
AT387464B (en) * 1982-10-29 1989-01-25 Voest Alpine Ag ARRANGEMENT FOR MONITORING THE LEVEL OF DAMAGE DUE TO MATERIAL EXHAUST IN CONSTRUCTIONS, VEHICLES, MACHINES AND THE LIKE
FR2539874B1 (en) * 1983-01-20 1985-07-05 Alsthom Atlantique MONITORING SYSTEM FOR TORSIONAL DAMAGE OF A SHAFT LINE COMPOSED OF A DRIVING MACHINE AND A DRIVEN MACHINE
IT1160818B (en) * 1983-03-22 1987-03-11 Rockwell Rimoldi Spa INSTRUMENT FOR THE PROCESSING OF DATA HAND AND DETECTED WITH AUTOMATIC EXECUTION OF THE CHECK CARD FOR AVERAGE (X) AND EXCURSION (W)
NO840900L (en) * 1983-04-14 1984-10-15 Carrier Corp TEMPERATURE MEASURES AND METHOD FOR AA STORING DATA
DE3314181A1 (en) * 1983-04-19 1984-10-25 Kraftwerk Union AG, 4330 Mülheim METHOD FOR MONITORING THE FATIGUE OF COMPONENTS, e.g. IN NUCLEAR POWER PLANTS
US4612620A (en) * 1983-06-06 1986-09-16 Ird Mechanalysis, Inc. Apparatus for collecting scheduled maintenance data
US4604706A (en) * 1983-06-27 1986-08-05 Integrated Sciences, Inc. Apparatus for failure prediction of earth structures
SE8303785L (en) * 1983-07-01 1985-01-02 Jan Ludwik Liszka SYSTEM FOR OPERATING CONTROL OF A MACHINE
JPS6021423A (en) * 1983-07-15 1985-02-02 Mitsubishi Electric Corp Vibration monitoring apparatus
US4611290A (en) * 1983-10-21 1986-09-09 Firetrol, Inc. Computer controlled diesel engine fire pump controller
JPH0614393B2 (en) * 1983-11-01 1994-02-23 能美防災株式会社 Fire detector
US4609994A (en) * 1984-01-16 1986-09-02 The University Of Manitoba Apparatus for continuous long-term monitoring of acoustic emission
JPS6139194A (en) * 1984-07-31 1986-02-25 ホーチキ株式会社 Fire alarm
ZA861082B (en) * 1985-04-24 1986-09-24 Dresser Ind Data monitor
DE3515061A1 (en) * 1985-04-26 1986-10-30 Fried. Krupp Gmbh, 4300 Essen METHOD AND DEVICE FOR MONITORING MACHINE PARTS
US4910692A (en) * 1985-10-09 1990-03-20 Outram John D Adaptive data logger
DE3607141A1 (en) * 1986-03-05 1987-09-10 Irs Ind Rationalis Syst Gmbh METHOD AND DEVICE FOR EXPLOSION PROTECTION OF SYSTEMS, PIPELINES AND THE LIKE THROUGH PRESSURE MONITORING
CN1040467C (en) * 1986-04-15 1998-10-28 Tlv有限公司 Steam trap operation judging device
FR2598524B1 (en) * 1986-05-07 1988-12-02 Berruyer Yves MONITORING METHOD FOR ANTICIPATING THE TRIGGERING OF AN ALARM.
US5005142A (en) * 1987-01-30 1991-04-02 Westinghouse Electric Corp. Smart sensor system for diagnostic monitoring
US4885707A (en) * 1987-02-19 1989-12-05 Dli Corporation Vibration data collecting and processing apparatus and method
KR890007306A (en) * 1987-10-30 1989-06-19 제트.엘.더머 Online valve diagnostic monitoring system
US4926342A (en) * 1987-12-31 1990-05-15 Westinghouse Electric Corp. High pressure rotor stress damage accumulating method
US4980844A (en) * 1988-05-27 1990-12-25 Victor Demjanenko Method and apparatus for diagnosing the state of a machine
US5251151A (en) * 1988-05-27 1993-10-05 Research Foundation Of State Univ. Of N.Y. Method and apparatus for diagnosing the state of a machine
US4935195A (en) * 1988-08-29 1990-06-19 Westinghouse Electric Corp. Corrosion-erosion trend monitoring and diagnostic system
US5099430A (en) * 1988-10-28 1992-03-24 Joseph Hirsch Method and apparatus for continuously suppressing unwanted rotational phenomena in a rotating body
GB8902645D0 (en) * 1989-02-07 1989-03-30 Smiths Industries Plc Monitoring
JP2574892B2 (en) * 1989-02-15 1997-01-22 株式会社日立製作所 Load sharing control method for automobile
AU6076690A (en) * 1989-07-17 1991-02-22 Williams Technology, Inc. Monitoring and control of oil/gas fields
EP0495001B1 (en) * 1989-10-02 1999-02-17 Rosemount Inc. Field-mounted control unit
US5335186A (en) * 1990-03-30 1994-08-02 Texas Instruments Incorporated Intelligent programmable sensing
US5094107A (en) * 1990-08-21 1992-03-10 The Minster Machine Company Press vibration severity/reliability monitoring system and method
US5239468A (en) * 1990-12-07 1993-08-24 United Technologies Corporation Automated helicopter maintenance monitoring
US5414365A (en) * 1992-09-25 1995-05-09 Martin Marietta Corporation Diagnostic apparatus for testing an analog circuit
EP0684586A1 (en) * 1994-05-27 1995-11-29 Tovarischestvo S Ogranichennoi Otvetstvennostju Finansovo-Promyshlennaya Korporatsia "Ljukon" (Too Fpk "Ljukon") Object condition diagnosing device containing an electronic operation parameter monitoring system
US5544073A (en) * 1994-06-02 1996-08-06 Computational Systems, Inc. Rotor balancing calculator
US5561610A (en) * 1994-06-30 1996-10-01 Caterpillar Inc. Method and apparatus for indicating a fault condition
US5675505A (en) * 1995-07-10 1997-10-07 Chrysler Corporation Sine on random data analysis method for simulating engine vibration
US5726911A (en) * 1996-08-22 1998-03-10 Csi Technology, Inc. Electric motor monitor
EP0891536B1 (en) 1996-12-12 2003-08-13 Prüftechnik Dieter Busch Ag System for monitoring units excited by oscillations
US6013108A (en) * 1997-03-18 2000-01-11 Endevco Corporation Intelligent sensor system with network bus
DE59712546D1 (en) * 1997-07-31 2006-04-06 Sulzer Markets & Technology Ag Method for monitoring systems with mechanical components
US5995910A (en) * 1997-08-29 1999-11-30 Reliance Electric Industrial Company Method and system for synthesizing vibration data
US6167318A (en) * 1997-12-22 2000-12-26 Alemite Corporation Oil mist generating system and method
US6351714B1 (en) 1998-03-03 2002-02-26 Entek Ird International Corporation Order tracking signal sampling process
US6363332B1 (en) * 1998-12-22 2002-03-26 Caterpillar Inc. Method and apparatus for predicting a fault condition using non-linear curve fitting techniques
US6195621B1 (en) * 1999-02-09 2001-02-27 Roger L. Bottomfield Non-invasive system and method for diagnosing potential malfunctions of semiconductor equipment components
GB2354825A (en) * 1999-07-30 2001-04-04 Eric Atherton Plant condition monitoring using vibrational measurements
GB2354606A (en) * 1999-09-22 2001-03-28 Martin Joseph Finnerty Predictive maintenance system
US6789030B1 (en) 2000-06-23 2004-09-07 Bently Nevada, Llc Portable data collector and analyzer: apparatus and method
GB0107900D0 (en) * 2001-03-29 2001-05-23 Post Office Improvements in monitoring systems
US6915237B2 (en) * 2002-05-14 2005-07-05 Analysis And Measurement Services Corporation Integrated system for verifying the performance and health of instruments and processes
US7254520B2 (en) 2002-05-14 2007-08-07 Analysis And Measurement Services Corporation Testing of wire systems and end devices installed in industrial processes
CN102032000B (en) * 2010-11-09 2013-08-28 浙江大学 Turbine status data acquisition device based on personal computer 104 (PC104) bus
US9451368B2 (en) * 2012-04-11 2016-09-20 Envoy Medical Corporation Feedback scan for hearing aid
US10301042B2 (en) * 2017-04-11 2019-05-28 General Electric Company Pitot-static air data test system with automatic dependent surveillance-broadcast (ADS-B) verification
CN110794486B (en) * 2018-08-02 2022-04-05 北京交通大学 Rainfall sensor

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3144774A (en) * 1961-02-14 1964-08-18 Stewart Warner Corp Vibration detection apparatus
US3699806A (en) * 1967-07-14 1972-10-24 Bjorn Weichbrodt Early detection of damage to machine elements in rolling engagement
US3731526A (en) * 1971-08-05 1973-05-08 United Aircraft Corp Variable center frequency filter
US3787670A (en) * 1971-12-20 1974-01-22 Honeywell Inf Systems Automatic diagnostic system
US3758758A (en) * 1972-03-10 1973-09-11 United Aircraft Corp Weighting rotational statistical vibration analysis employing sidebands and/or
US3763397A (en) * 1972-06-15 1973-10-02 Phillips Petroleum Co Monitoring and shutdown apparatus
US3783680A (en) * 1972-07-12 1974-01-08 Atomic Energy Commission Multipoint vibration monitoring system
US3882305A (en) * 1974-01-15 1975-05-06 Kearney & Trecker Corp Diagnostic communication system for computer controlled machine tools
DE2439072A1 (en) * 1974-08-14 1976-02-26 Hofmann Maschf Geb METHOD AND DEVICE FOR THE SELECTIVE MEASUREMENT OF VIBRATION COMPONENTS OF A VIBRATION SPECTRUM, IN PARTICULAR FOR IMBALANCE MEASUREMENTS, AND MEASURING VALUES FOR THIS
US4060716A (en) * 1975-05-19 1977-11-29 Rockwell International Corporation Method and apparatus for automatic abnormal events monitor in operating plants
US4010637A (en) * 1975-10-06 1977-03-08 Lockheed Aircraft Corporation Machine vibration monitor and method

Also Published As

Publication number Publication date
US4184205A (en) 1980-01-15
EP0002232A3 (en) 1980-01-23
DE2862285D1 (en) 1983-07-28
CA1139881A (en) 1983-01-18
EP0002232A2 (en) 1979-06-13
JPS5494061A (en) 1979-07-25

Similar Documents

Publication Publication Date Title
EP0002232B1 (en) System and method for monitoring the operation of an apparatus
USRE31750E (en) Data acquisition system
US4060716A (en) Method and apparatus for automatic abnormal events monitor in operating plants
US4887468A (en) Nonsynchronous turbine blade vibration monitoring system
US5210704A (en) System for prognosis and diagnostics of failure and wearout monitoring and for prediction of life expectancy of helicopter gearboxes and other rotating equipment
US5710723A (en) Method and apparatus for performing pre-emptive maintenance on operating equipment
US5852793A (en) Method and apparatus for predictive diagnosis of moving machine parts
EP0055297A1 (en) Monitoring device for a rotary machine
US20090037121A1 (en) System and method for detection of rotor eccentricity baseline shift
US7444265B2 (en) Machine and/or monitoring
US4136396A (en) Data processing
US6873918B2 (en) Control embedded machine condition monitor
CN114813124B (en) Bearing fault monitoring method and device
US20040193387A1 (en) Signal recorder with status recognizing function
US7249287B2 (en) Methods and apparatus for providing alarm notification
GB2122749A (en) Electrical condition monitoring of electric motors
JPH07261838A (en) Method and device for monitoring performance of plant equipment or system
US4218734A (en) Process quantity display apparatus
JPH02232529A (en) Method and apparatus for diagnosing vibration of rotary machine
JPH06274778A (en) Plant monitoring/diagnosing device and its fault sign detecting method
GB2175116A (en) Fault recognition equipment for machine operation
KR100496126B1 (en) coil current monitoring system for control element drive mechanism control system and methods thereof
EP0715738A4 (en) Interpretive measurement instrument
Hills On-line Surveillance for Predictive Maintenance in a Hazardous Environment
JP2001255243A (en) Abnormality monitoring system of rotating equipment

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Designated state(s): BE DE FR GB NL

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AK Designated contracting states

Designated state(s): BE DE FR GB NL

17P Request for examination filed
GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Designated state(s): BE DE FR GB NL

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: NL

Effective date: 19830622

Ref country code: BE

Effective date: 19830622

REF Corresponds to:

Ref document number: 2862285

Country of ref document: DE

Date of ref document: 19830728

ET Fr: translation filed
NLV1 Nl: lapsed or annulled due to failure to fulfill the requirements of art. 29p and 29m of the patents act
PLBI Opposition filed

Free format text: ORIGINAL CODE: 0009260

26 Opposition filed

Opponent name: KRAFTWERK UNION AKTIENGESELLSCHAFT

Effective date: 19840322

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 19841031

Year of fee payment: 7

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 19841108

Year of fee payment: 7

PLBN Opposition rejected

Free format text: ORIGINAL CODE: 0009273

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: OPPOSITION REJECTED

27O Opposition rejected

Effective date: 19850415

GBPC Gb: european patent ceased through non-payment of renewal fee
PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 19880729

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Effective date: 19880802

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Effective date: 19881117

PLAB Opposition data, opponent's data or that of the opponent's representative modified

Free format text: ORIGINAL CODE: 0009299OPPO